This is the mail archive of the java-patches@sourceware.cygnus.com mailing list for the Java project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Patch: fix -ms, -mx options for gij


The options to set initial, maximum heap did not work as advertised. 
I've tested the patch below.  It is _not_ committed.

Jeff


1999-11-05  Jeff Sturm  <jsturm@sigma6.com>

        * boehm.cc (_Jv_GCSetInitialHeapSize): Swapped size & current.
        * prims.cc (parse_heap_size): Use end, not spec.  Use 1024
        multipler for `k'.

Index: boehm.cc
===================================================================
RCS file: /cvs/java/libgcj/libjava/boehm.cc,v
retrieving revision 1.10
diff -u -r1.10 boehm.cc
--- boehm.cc    1999/11/03 01:15:40     1.10
+++ boehm.cc    1999/11/05 17:11:01
@@ -375,7 +375,7 @@
 {
   size_t current = GC_get_heap_size ();
   if (size > current)
-    GC_expand_hp (current - size);
+    GC_expand_hp (size - current);
 }

 void
Index: prims.cc
===================================================================
RCS file: /cvs/java/libgcj/libjava/prims.cc,v
retrieving revision 1.12
diff -u -r1.12 prims.cc
--- prims.cc    1999/10/22 19:43:39     1.12
+++ prims.cc    1999/11/05 17:11:02
@@ -822,10 +822,10 @@
 {
   char *end;
   unsigned long val = strtoul (spec, &end, 10);
-  if (*spec == 'k' || *spec == 'K')
-    val *= 1000;
-  else if (*spec == 'm' || *spec == 'M')
-    val *= 1000000;
+  if (*end == 'k' || *end == 'K')
+    val *= 1024;
+  else if (*end == 'm' || *end == 'M')
+    val *= 1048576;
   return (size_t) val;
 }

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]